The specific aims of the proposed study are to: (a) test a social cognitive-based intervention designed to promote sexual abstinence and the use of HIV risk-reduction practices among adolescent males; (b) to enhance the father s role in promoting sexual abstinence and HIV prevention behaviors among their sons; and c) to test the mediators of abstinence for the intervention. The Boys and Girls Clubs of Metro Atlanta will serve as sites for the study. Adolescent males and their fathers will first complete a baseline interview, and then will be assigned together to one of two conditions: Condition 1 (Social Cognitive Treatment) consists of a social cognitive-based intervention aimed at enhancing the father s role in promoting sexual health of their sons. Condition 2 (CONTROL) consists of a Health Promotion Program. Fathers only will attend the 7 weekly intervention/program sessions. Fathers will receive instruction and materials so that they can teach their sons the information they are receiving in the program sessions. Follow-up interviews will be completed by both fathers and their sons at 3, 6, and 12 months. Participants will be adolescent boys ages 11-14 years who are recruited through the Boys and Girls Clubs and their fathers (or male guardians). The total time to recruit participants and conduct the interventions is expected to be 20 months (following a 6-month start-up period). Each participant will be followed for 12 months. A nested cohort design with sites rather than individuals randomly assigned to conditions will be employed with participants nested within sites. With the effects of randomization, intracluster correlations, and attribution incorporated into the sample size calculations, we estimate that we will need 156 fathers and 156 adolescent male participants per conditions (total 312 fathers and 312 sons). To test the primary hypothesis that a greater proportion of the adolescent males at baseline who are exposed to the social cognitive intervention will be more likely to remain abstinent at posttestings than those in the control group, a test of group difference in proportion will be used. Statistical tests will also be conducted to assess proposed mediators of the intervention.